charlieobaugh.com — AI Site Grade

Analysis

Charlie Obaugh Auto Group — AI-Visibility Audit

The site's robots.txt explicitly blocks GPTBot, ClaudeBot, Bytespider, Google-Extended, Applebot-Extended, CCBot, and Amazonbot via Cloudflare-managed rules, yet Cloudflare enforces those blocks inconsistently — GPTBot, ClaudeBot, and Bytespider receive HTTP 403 at the edge, while Google-Extended and Applebot-Extended get HTTP 200 with full content, creating a fragmented AI-crawler posture where some training bots are walled off but search-augmentation bots pass through.

Crawler Access

The robots.txt uses Cloudflare's Content-Signal framework (search=yes, ai-train=no) for the wildcard rule, then individually disallows every major AI crawler. In practice, compare_bot_access shows GPTBot, ClaudeBot, OAI-SearchBot, ChatGPT-User, PerplexityBot, Perplexity-User, Bytespider, and anthropic-ai all return 403 (Cloudflare block page, 25 bytes). Google-Extended and Applebot-Extended return 200 with ~244KB of content matching the browser baseline. The site runs on Cloudflare + Next.js (hosted via Alpha Dealer Sites), with cache-control: private, no-cache, no-store — meaning even permitted bots get uncached, dynamically rendered pages. The /llms.txt returns a 404 (Next.js error page with noindex meta tag). The sitemap.xml contains 1,278 URLs covering inventory, brands, and service pages.

Cold-Knowledge Gap

A frontier LLM queried cold about "Charlie O Baugh" returned zero knowledge — could not confirm the brand, its products, location, or industry. The site itself is a multi-franchise auto group in Staunton, VA selling Chevrolet, GMC, Kia, Mitsubishi, and RV/outdoor vehicles across four physical locations. The gap between the model's blank slate and the site's actual content (415 new vehicles, 146 used, 64 active specials, service centers, finance) is total. AI engines with no retrieval access to this domain will describe the brand as unknown, which is a catastrophic visibility failure for a dealership competing for local search and AI-generated answers.

Schema Posture

Every page examined — homepage, about, service, finance, specials, new inventory — contains zero JSON-LD schema of any type. No AutoDealer, LocalBusiness, Product, Vehicle, FAQPage, or Organization markup is present. The homepage has no structured data for the four physical locations, no OpeningHoursSpecification, no AggregateRating despite displaying "10K+ Customers" and "98% Satisfaction" in the UI. The about page lists staff (Charlie Obaugh, Eric Obaugh, Ben Puckett) with contact info but no Person or EmployeeRole schema. This is a complete structured-data vacuum.

External Signals

Web searches for "Charlie Obaugh Auto Group", "Charlie Obaugh Staunton", and "Charlie Obaugh reviews" across DuckDuckGo returned zero results — no press, no Reddit threads, no review-site citations surfaced. The site's external links point to brand-specific subdomains (charlieobaughgm.com, charlieobaughkia.com, charlieobaughmitsubishi.com, charlieobaughrv.com), all of which return 403 when fetched directly, suggesting they are separate dealer-platform instances behind Cloudflare. The only external link in the footer is to alphadealersites.com, the platform provider. The dealership has no detectable off-domain footprint that AI engines could cite.

Additional Findings

The homepage and inventory pages render 903 and 830 words of visible text respectively from a plain GET — the site is not a JS shell, so content is accessible to bots that are not blocked. However, the robots.txt Content-Signal ai-train=no combined with individual bot disallows creates a contradictory signal: the wildcard says "allow /" while the bot-specific rules say "disallow /". The cache-control: private, no-cache header prevents CDN caching even for permitted crawlers, forcing every AI bot request to hit the origin server. The finance page has only 132 words of content. No FAQ, comparison, or table answer-format signals exist anywhere on the site.

Findings

1. Robots.txt blocks most AI training crawlers High

   The robots.txt disallows GPTBot, ClaudeBot, Bytespider, Google-Extended, Applebot-Extended, CCBot, and Amazonbot. Cloudflare enforces blocks inconsistently: GPTBot, ClaudeBot, and Bytespider receive HTTP 403, while Google-Extended and Applebot-Extended get HTTP 200 with full content.

   What to change: Remove disallow rules for AI crawlers that should be allowed, or adjust Cloudflare settings to consistently allow desired bots.

2. Zero JSON-LD schema markup on any page High

   Every page examined—homepage, about, service, finance, specials, new inventory—contains no JSON-LD schema of any type. No AutoDealer, LocalBusiness, Product, Vehicle, or Organization markup is present.

   What to change: Add JSON-LD structured data for AutoDealer, LocalBusiness, Vehicle, Product, and Organization schemas on relevant pages.

3. LLM cold knowledge returns zero information about the brand High

   A frontier LLM queried about 'Charlie O Baugh' returned no knowledge—could not confirm the brand, products, location, or industry. The site is a multi-franchise auto group in Staunton, VA with 415 new vehicles, 146 used, and 64 specials.

   What to change: Allow AI crawlers access to the site and add structured data to help LLMs understand the business.

4. No external citations found in web searches High

   Web searches for 'Charlie Obaugh Auto Group', 'Charlie Obaugh Staunton', and 'Charlie Obaugh reviews' returned zero results across DuckDuckGo. No press, Reddit threads, or review-site citations were found.

   What to change: Build external citations through press releases, local listings, and social media to create off-domain signals.

5. /llms.txt returns 404 Medium

   The /llms.txt endpoint returns a 404 Next.js error page with a noindex meta tag, providing no machine-readable summary for AI assistants.

   What to change: Create an llms.txt file with a summary of the site's content and key pages for AI assistants.

6. Cache-control headers prevent caching for permitted bots Medium

   The site sets cache-control: private, no-cache, no-store, meaning even permitted AI crawlers receive uncached, dynamically rendered pages, increasing server load and latency.

   What to change: Adjust cache-control headers to allow caching for permitted crawlers, e.g., public, s-maxage=3600.

7. Contradictory robots.txt signals for AI crawlers Medium

   The wildcard rule uses Cloudflare's Content-Signal 'search=yes, ai-train=no' while individual bot rules disallow '/', creating a contradictory signal that may confuse crawlers.

   What to change: Simplify robots.txt to either allow or disallow AI crawlers consistently, avoiding mixed signals.

8. Thin content on finance and service pages Medium

   The finance page contains only 132 words and the service page 174 words, providing minimal information for AI to understand these offerings.

   What to change: Expand content on finance and service pages to at least 500 words each, including details on financing options and services offered.

9. No FAQ or answer-format signals on the site Medium

   The site lacks FAQPage schema, comparison tables, or any structured answer formats that AI assistants use to generate direct answers.

   What to change: Add FAQPage schema to pages with common questions, and consider adding comparison tables for vehicle models.

10. Brand subdomains return 403 errors Medium

    Subdomains like charlieobaughgm.com, charlieobaughkia.com, charlieobaughmitsubishi.com, and charlieobaughrv.com return HTTP 403, making them inaccessible to AI crawlers.

    What to change: Ensure subdomains are accessible to AI crawlers or consolidate content onto the main domain.

What's working

• Content is accessible to permitted bots — The homepage and inventory pages render 903 and 830 words of visible text respectively from a plain GET, meaning content is not hidden behind JavaScript and is accessible to bots that are not blocked.
• Sitemap present with 1,278 URLs — The sitemap.xml contains 1,278 URLs covering inventory, brands, and service pages, providing a comprehensive map for crawlers.
• Google-Extended and Applebot-Extended allowed — Despite blocking many AI crawlers, Google-Extended and Applebot-Extended receive HTTP 200 with full content, enabling search-augmentation AI to access the site.